Thursday 23 July 2020

Gyotaku Fish rubbings as a source of historical biodiversity data.

Access to historical biodiversity information is limited, being mainly obtained via museum specimens, past literature, movies, photographs/images and/or other historical materials such as classic monographs. These potential data sources, in particular historical materials, are sometimes lost by deliberate or accidental disposal or, for example, by fire or an estate liquidation where materials are scattered far from their origin. There are other similar examples where biodiversity information can be lost over time: seeds in seed banks can die within decades of their collection, and environmental DNA cannot be detected several hours or days after sampling. Thus, it is important for historical biodiversity information to be accessed and recorded as a matter of urgency. Copying or extracting information from privately owned materials is of highest priority. Biodiversity observations have been made not only by researchers but also by citizens, even prior to the recent rise of citizen science projects. Although data mining from citizens’ observation records is a legitimate method of citizen science, limitations on data availability have not been well documented except for online data. In Japan, many recreational fishers have recorded their memorable catches as ‘Gyotaku’ (魚拓), which means 'Fish impression' or 'Fish rubbing' in English. since the last Edo Period (the current oldest known Gyotaku dates back to late February 1839). Gyotaku is made directly from Fish specimen(s), and usually includes information such as sampling date and locality, the name(s) of the fisher(s), its witness(es), the Fish species (frequently its local name), and fishing tackle used. In recent decades, color versions of Gyotaku have become well developed, and used for art and educational purposes. In contrast, the traditional method is printed by using black writing ink. Generally, color prints for art rarely include specimen data including sampling locality and date.

In a paper published in the journal ZooKeys on 16 January 2020, Yusuke Miyazaki of the Department of Child Studies and Welfare at Shiraume Gakuen College, and Atsunobu Murase of the Nobeoka Marine Science Station and Department of Marine Biology and Environmental Sciences of the University of Miyazaki, attempt to validate the hypotheses that historical biodiversity data attached to Gyotaku prints are at risk of being lost, and that the number of Gyotaku prints is generally declining, being replaced with photographs from digital cameras and/or smart phones.

An explanation of Fish rubbing (Gyotaku, in Japanese). (A) First step, the Fish specimen is painted using ink. (B) Second step, the specimen is covered with a sheet of paper. (C) The finished image of the Fish specimen on the paper. This is known as the direct method of Gyotaku; there is also an indirect method whereby a sheet of paper is placed on the Fish specimen, then the sheet is painted by hand using ink. Miyazaki & Murase (2020).

The number of fishing-related shops that are personally managed (rather than the large chain stores) and therefore likely to stock original Gyotaku prints may be decreasing in recent years due to their owners retiring, an increase in chain store numbers, and/or a decrease in recreational fishers. Miyazaki and Murase attempt to validate these hypotheses by collecting data of Gyotaku from recreational fishing shops where threatened fish species are distributed according to both the national and regional Red Lists. The potential use of Gyotaku for historical biodiversity information is discussed.

Preliminary field surveys were conducted at three fishing shops in Miyazaki Prefecture, and one recreational boating shop in Chiba Prefecture where Miyazaki and Murase found Gyotaku of threatened species were stocked via research and by chance during in August 2016 (Miyazaki Prefecture), and in March 2017 (Chiba Prefecture). They also conducted preliminary surveys of recreational fishing shops at the northernmost and southernmost regions of Japan in order to understand the Gyotaku information available at the latitudinal limits of Japan (from the subarctic to the tropics). Referencing the Town Page (the yellow pages of the Nippon Telegraph and Telephone Corporation) of the Souya (northernmost) and the Yaeyama (southernmost) regions, Miyazaki and Murase identified relevant shops (six for Souya in November 2017, and ten for Yaeyama in May 2017). In these surveys, we asked for information on the presence/absence of Gyotaku and the possibility of photographing them, and other relevant data. Where possible, Miyazaki and Murase photographed Gyotaku stocks, and sought permission to use the images for research.

Miyazaki and Murase's questionnaires mainly surveyed three regions of Japan where threatened Fish species are distributed according to the national Red List: the Sakhalin Taimen, Hucho perryi, for Hokkaido (234 shops and stores); the Small-scale Sillago, Sillago parvisquamis, for Tokyo Bay (274 shops and stores); and the Japanese Lates, Lates japonicus, for Miyazaki Prefecture (80 shops and stores). The Souya region, which had been previously surveyed and no Gyotaku with distributional data were recorded from there, was excluded from this.

An explanation of the aim of Miyazaki and Murase's research and an answer sheet, which covered fifteen items for informed consent based on a research ethics review at Shiraume Gakuen College, were attached to the questionnaire. In the surveys conducted during July–September 2018, Miyazaki and Murase asked for information on the presence/absence of Gyotaku, and the possibility of copying relevant data. When possible, they conducted field surveys to photograph Gyotaku in March 2019.

Of the stores and shops targeted by our second preliminary surveys, none (of six) in the northernmost (Souya) region and three (of ten) in the southernmost (Yaeyama) region stocked Gyotaku rubbings with distributional information.

Regarding the questionnaires, fourteen surveys were returned unopened due to stores being closed down, and 56 responses were received from others, indicating that the questionnaire response rate was 9.5%. Miyazaki and Murase's field surveys were permitted by nine stores and shops that stocked Gyotaku rubbings, based on the questionnaire surveys, while 82% of the responses recorded no stock of Gyotaku. This low response rate was possibly caused by us not paying to have the surveys completed and by a high perceived workload to complete the answers.

In total, 261 Gyotaku rubbings, with 325 printed individual specimens (i.e., a part of Gyotaku has multiple individuals on a single sheet), were found among the targeted shops All data recorded were integrated to the Gyotaku Database. Among the data obtained in Miyazaki and Murase's study, they extracted distributional data for 221 individuals. Distributional data for an additional 14 individuals were obtained through interviewing the holders of the Gyotaku regarding date and/or locality information, resulting in a total of 235 individuals with distributional data. Among the prints, 68 Japanese Fish and three Cephalopod species were represented, but 65 of these (14.9%) did not include a Fish name. In general, a limited number of species are targeted by recreational fishers. Anglers and shop staff accurately identify the main fishing targets, and misidentifications are quickly corrected by other fishers in the local community. A pertinent example is the Common Octopus, Octopus vulgaris, which is identified as a main fishing target by one fishing shop in Yokohama City, Tokyo Bay. Gyotaku images of an Octopus from this shop is likely to be Octopus vulgaris even if there is no name on the specific Gyotaku. However, this study did not validate such identifications based on external morphology and/or molecular analyses.

The observed species compositions reflected the biogeography of the regions. For example, prints of seven individuals of Hucho perryi were recorded from only Hokkaido. Similarly, one individual of Sillago parvisquamis was recorded from only around Tokyo Bay, while three individuals of Lates japonicus were recorded from only Miyazaki Prefecture. These three species are listed as threatened in national and prefectural Red Lists. In particular, populations of Sillago parvisquamis are probably extinct in Tokyo Bay. The last reliable record from Tokyo Bay is from 1975–1976. Additionally, the populations of Lates japonicus at Miyazaki Prefecture were listed on the Specified Prefectural Endangered Species of Wild Fauna and Flora on 21 December 2012; this prohibits the capture, holding, receiving, and giving of, and other interactions with, the species without the prefectural governor’s permission. Given the rarity of these threatened species in some regions, ‘gyotaku’ are probably important vouchers for estimating historical population status, and factors of decline or extinction.

Three species targeted in leisure fishing and listed as threatened species in the Japanese national Red List. (A) Hucho perryi from a shop in Hokkaido. (B) Sillago parvisquamis from a shop facing Tokyo Bay. (C) Lates japonicus from a shop in Miyazaki Prefecture. Miyazaki & Murase (2020).

Species belonging to other families such as Salmonidae and Pleuronectidae, which originate in cold waters, were mostly recorded from Hokkaido rather than the other surveyed regions. Furthermore, several Carangid Fish, Caranx spp., the Okinawa Seabream, Acanthopagrus sivicolus, the Spangled Emperor, Lethrinus nebulosus, the Oranges-potted Spinefoot, Siganus guttatus,and others originating from warm waters were recorded from only the Yaeyama region. Another Seabream species, Acanthopagrus latus, which shares a similar distributional pattern with Lates japonicus, was recorded from only Miyazaki Prefecture. Moreover, Miyazaki and Murase's list also included exotic non-native species, i.e. the Rainbow Trout, Oncorhynchus mykiss, and the Brown Trout, Salmo trutta, from the fishing tackle stores of Hokkaido, and Oncorhynchus mykiss and the Largemouth Bass, Micropterus salmoides, from a fishing tackle store in Tokyo Metropolis. Furthermore, some images printed of Gyotaku were found at stores well outside the pictured species’ known range; generally, these resources had been provided by customers who had traveled to other regions for leisure fishing trips. Overall, the species composition displayed in the Gyotaku approximately reflected the Fish faunas of each biogeographic region.

Miyazaki and Murase also obtained a statistically estimated result using a state space model. This estimation showed very few Gyotaku available from before 1985, with a peak in 2002. These results suggest that using this technique to gather historical data is valid for perhaps the last 30 years or so and not prior to that. Obtaining useful Gyotaku more than 30 years old is unlikely. A decline in number was observed during 2011 and 2012, which probably reflects an indirect effect of the catastrophic tsunamis and the nuclear accidents caused by the Great East Japan Earthquake on 11 March 2011. Miyazaki and Murase's data does not support the hypothesis that the use of Gyotaku will be decreasing over recent years due to the rise of digital photography. They suggest that Japanese recreational fishers may be continuing to use the Gyotaku method in addition to digital photography to record their memorable catches.

Currently, the oldest Gyotaku material is a collection of the Tsuruoka City Library made in 1839. Others from the 19th Century were made in 1850s–1860s and are now collections of the Homma Museum of Art and the Chido Museum. The oldest material found in Miyazaki and Murase's study was made in 1936, indicating that it is difficult to find very old Gyotaku rubbings at leisure fishing stores and shops. Storage of Gyotaku in the public areas of shops and stores is usually less than ideal, with exposure to tobacco smoke, sunlight, and moisture. This is the main reason for deteriorating Gyotaku. In fact, some shop owners reported disposing of older damaged materials. Further field surveys of, for example, museum and private collections are required to discover older Gyotaku and extract relevant data.

In conclusion, distributional data related to Fish diversity records were able to be mined from Gyotaku. However, this method is time limited with respect to data rescue from the general public. The volume of data obtained in this study is too small to analyse statistically from the perspective of ecology, biogeography, or other similar disciplines. Additionally, validation of the identifications sourced from the Gyotaku is required via taxonomic evaluations. This could be done by examining the external morphology captured in the printed image and possibly by trying to obtain biological material from the print for molecular analysis (i.e., based on the residuum of dried DNA on the sheet). Overall, further research is required into the use of Gyotaku rubbings for acquiring historical biodiversity data.

See also...
Follow Sciency Thoughts on Facebook.